摘要
为提高支持向量机在建模方面的拟合性能,针对核函数方法中单个核函数的局限性,尝试融合核支持向量机建模方法以提高模型的泛化能力和精度。为避免在进行核融合时,支持向量机稀疏性的缺失,提出将数据映射到稀疏特征空间进行研究。通过仿真研究表明,所建模型在保证稀疏性的前提下,能较好地提高建模精度,从而验证了算法的有效性。
In order to improve SVM＇s fitting performance of modeling, a method based on fusion kernels was tried to improve the generalization ability and the precision of models, according to the limitation of single kernel support vector machine. In research, sample data were mapped to sparse feature space to prevent the loss of SVM＇ s sparsity when the kernels were fused. Through the simulation, the results show that the modeling method can improve the modeling accuracy in the premise of keeping sparsity, thus providing the method＇s effectiveness and applied meaning.
出处
《化工自动化及仪表》
CAS
北大核心
2009年第4期30-32,37,共4页
Control and Instruments in Chemical Industry
基金
"863"国家高科技计划资助项目(2006AA020301)
关键词
核函数
支持向量机
稀疏特征空间
建模
kernel function
support vector machine
sparse feature space
modeling
